Inhibitory Mechanisms of Tea Polyphenols on the Ultraviolet B-activated Phosphatidylinositol 3-Kinase-dependent Pathway*

In this study, we investigated the effect of tea polyphenols, ( (cid:1) )-epigallocatechin-3-gallate or theaflavins, on UVB-induced phosphatidylinositol 3-kinase (PI3K) activation in mouse epidermal JB6 Cl 41 cells. Pretreatment of cells with these polyphenols inhibited UVB-in-duced PI3K activation. Furthermore, UVB-induced activation of Akt and ribosomal p70 S6 kinase (p70 S6-K), PI3K downstream effectors, were also attenuated by the polyphenols. In addition to LY294002, a PI3K inhibitor, pretreatment with a specific mitogen-activated protein/ extracellular signal-regulated protein kinases (Erks) kinase 1 inhibitor, U0126, or a specific p38 kinase inhibitor, SB202190, blocked UVB-induced activation of both Akt and p70 S6-K. Pretreatment with LY294002 restrained UVB-induced phosphorylation of Erks, suggesting that in UVB signaling, the Erk pathway is mediated by PI3K. Moreover, pretreatment

Nonmelanoma skin cancer is the most frequently diagnosed malignancy in the United States, and the most crucial risk factor in development of nonmelanoma skin cancer is UV radiation from sunlight (1,2). Although the UV portion of sunlight is divided into three components based on wavelength, the UV light that reaches the surface of the earth is comprised of UVA (320 -400 nm) and UVB (280 -320 nm). UVB is a major cause of skin cancer (3,4). In animal models, UV radiation acts as both a tumor initiator and tumor promoter (5,6). The initiating effects of UV radiation include chromosomal alterations and mutations that result in DNA damage (7,8). The promoting effects may result from activation of transcription factors that control the expression of genes involved in tumor promotion (7, 9 -12). The mitogen-activated protein (MAP) 1 kinase family (e.g. extracellular signal-regulated protein kinases (Erks), c-Jun N-terminal kinases (JNKs), and p38 kinase) at least partially mediates the activation of transcription factors (13)(14)(15)(16). In addition, UV radiation triggers activation of tyrosine receptors, such as the epidermal growth factor (EGF) receptor and the insulin receptor, Src family tyrosine kinases, and the small GTP-binding protein, Ras, and MAP kinases (11)(12)(13)(14)(17)(18)(19). On the other hand, UV radiation also causes activation of the phosphatidylinositol 3-kinase (PI3K) pathway as well as the activation of tyrosine receptors and Ras (20,21). A family of PI3K enzymes phosphorylates the number 3 position of the inositol ring of some different phosphoinositides (22). Among the phosphoinositides, phosphatidylinositol (4,5) diphosphate is believed to be the preferred substrate in vivo generating the second messenger phosphatidylinositol (3,4,5) triphosphate (23,24). PI3K plays a central role in a broad range of biological effects such as cell growth, apoptosis, intercellular vesicle trafficking/secretion, regulation of actin, cell migration, and integrin function (25,26). In addition, accumulating evidence suggests the importance of PI3K signaling in carcinogenesis (27,28). Initially PI3K was a subject of interest because of its known ability to form complexes with some viral oncoproteins (29,30) and also because of its involvement in the viral transformation process (23). Later, expression of a retroviral oncogene, v-p3k, which codes for the catalytic subunit (p110␣) of PI3K, was shown to induce oncogenic transformation in chicken cells (31). The oncogenic transformed phenotype was observed in mammalian fibroblasts transfected with the constitutively active p110␣ (32). In fact, alterations or amplifications of PI3K have been detected in a number of human malignances (33,34). Our previous studies also demonstrated that PI3K is necessary for 12-O-tetradecanoylphorbol-13-acetate-or EGF-induced cell transformation in mouse epidermal JB6 cells (35,36). Those findings suggested that PI3K activation by UV exposure is a critical factor in UV-induced carcinogenesis. The serine/threonine proto-oncogene Akt (also known as protein kinase B or PKB) was first identified as the cellular homolog of the transforming oncogene product v-Akt (37). It plays an im-portant role in the regulation of a wide spectrum of cellular signaling events including apoptosis, glycogen metabolism, protein synthesis, and cell cycle regulation (38,39). Akt is activated by growth factors and oncogenes through PI3K-dependent signaling (38,40). Akt has been shown to be overexpressed in ovarian, prostate, breast, and pancreatic cancers and is associated with a poor prognosis and increased tumorigenicity (27,(41)(42)(43). Alteration and down-regulation of the tumor suppressor gene PTEN/MMAC1, which directly antagonizes PI3K, are frequently observed in a number of cancers (27,44). The mutations are thought to be associated with Akt activation as well as an increased level of phosphatidylinositol 3,4,5-triphosphate (45,46). An accumulation of evidence suggests that ribosomal p70 S6 kinase (p70 S6-K) is also a crucial effector of PI3K in response to growth factors (47). Activation of p70 S6-K regulates a variety of cellular functions involved in the mitogenic response including protein synthesis, translation of specific mRNA species, cell cycle regulation, and cell motility (48 -50). The p70 S6-K is constitutively activated in small cell lung cancer cells, and anchorage-independent proliferation in these cells is mediated through an Akt and p70 S6-K-dependent pathway (51). Kl-Ras-mediated transformation is inhibited by blocking both the MAP kinase and p70 S6-K pathways (52). Activation of p70 S6-K has been shown to be essential for the UVB-induced increase in MMP-1 and MMP-3 proteins in human dermal fibroblasts (53).
Tea and tea polyphenol preparations have been shown to inhibit carcinogenesis at several organ sites in rodent models (54,55). The polyphenols from green tea and black tea, (Ϫ)epigallocatechin-3-gallate (EGCG) and theaflavins, respectively, are generally considered to be the most effective components for inhibition of carcinogenesis (56 -58). Tea polyphenols also have been shown to inhibit UV-induced carcinogenesis in animal models (56, 59 -61). The most widely recognized biological properties of tea polyphenols are their antioxidant activity in scavenging reactive oxygen or nitrogen species (62). The molecular mechanism of their anti-tumor promotion effects might be related to their ability to block signal transduction pathways leading to carcinogenesis (57,58,63). Recently, we have shown that tea polyphenols inhibit UVB-induced AP-1 activation in mouse epidermal JB6 cells (64). Thus, the inhibitory effect of the tea polyphenols in UVB-induced carcinogenesis may result from the blocking of these signal transduction pathways. In the present study, we investigated the effects of tea polyphenols on UVB-induced activation of the PI3K and its downstream effectors, which are critical factors in carcinogenesis.
UV Irradiation-UVB irradiation was performed on serum-starved monolayer cultures utilizing a transluminator emitting UVB (66). The source of UVB is a bank of four Westinghouse F520 Lamps (National Biological, Twinsburg, OH) at 6 J/S/m light in the UVB range. Approximately 10% of the remaining radiation from the F520 Lamp was in the UVA region (320 nm). Although almost no UVC leakage occurs, the UVB irradiation was carried out in a UVB exposure chamber fitted with a Kodak Kodacel K6808 filter that eliminates all wavelengths below 290 nm. This lamp is one of the most frequently used UVB sources for the study of carcinogenesis. The International Agency for Research on Cancer refers to this lamp as a source emitting mainly UVB irradiation for the studies of cancer induction in animals. UVB irradiation was measured using the UVX radiometer from UVP (UVX-31).
Cell Culture-The JB6 mouse epidermal cell line Cl 41 and its stable transfectants, Cl 41 CMV-neo and the dominant negative mutant of JNK1 were grown at 37°C in MEM supplemented with 5% heatinactivated FBS, 2 mM L-glutamine, and 25 g/ml gentamicin.
Generation of Stable Cotransfectants-The dominant negative mutant of JNK1 subcloned into a mammalian expression vector plasmid, CMV-neo, was transfected into JB6 Cl41 cells by using LipofectAMINE (Life Technologies Inc.) following the manufacturer's instructions. The stable transfectants were obtained by selection for G418 resistance (300 g/ml) and further confirmed by activity assays.
Immunoblotting-Immunoblotting was carried out as described previously (64). In brief, JB6 Cl 41 cells and its stable transfectants were cultured to 80% confluence. The cells were starved in 0.1% FBS MEM for 48 h at 37°C. Then the media were changed to fresh 0.1% FBS MEM, and the cells were incubated for another 2-4 h at 37°C. Before the cells were irradiated with UVB, they were incubated with or without LY294002, PD98059, SB202190, rapamycin, EGCG, or theaflavins for 1 h. Then the cells were exposed to UVB (4 kJ/m 2 ) and subsequently incubated for an additional 30 min at 37°C in the presence of inhibitors or tea polyphenols. The cells were then lysed, and immunoblot analysis was performed by using antibodies against total Akt, p70 S6-K, Erks, JNKs, or p38 kinase proteins or the phospho-specific antibodies against their phosphorylated proteins. Antibody-bound proteins were detected by chemiluminescence (ECF Western blotting kit; Amersham Pharmacia Biotech) and analyzed using the Storm 840 Scanner (Molecular Dynamics, Sunnyvale, CA).
Assay for JNK Activity-The cells were treated with UVB (4 kJ/m 2 ) as described above. The cells were lysed in 400 l of lysis buffer (20 mM Tris, pH 7.4, 150 mM NaCl, 1 mM Na 2 EDTA, 1 mM EGTA, 1 mM Na 3 VO 4 , 1 mm ␤-glycerophosphate, 1% Triton X-100, 2.5 mM sodium pyrophosphate, 1 mM phenylmethylsulfonyl fluoride, 1 g/ml leupeptin, and 1 M microcystin). The lysates were sonicated and centrifuged, and the supernatant fractions were incubated with a phospho-specific JNKs antibody with gentle rocking overnight at 4°C. Then, protein A/G plus agarose was added, and the incubation continued for another 4 h at 4°C. The beads were washed twice with 500 l of lysis buffer and twice with 500 l of kinase buffer (25 mM Tris, pH 7.5, 5 mM ␤-glycerolphosphate, 2 mM DTT, 0.1 mM Na 3 VO 4 , 10 mM MgCl 2 ). The kinase reactions were carried out in the presence of 200 M ATP at 30°C for 30 min using 2 g of c-Jun as substrate for JNKs. The phosphorylated protein was detected by immunoblotting using a phospho-specific antibody.
Akt and p70 S6-K Immunoprecipitation Kinase Assay-The cells were treated with the inhibitors or tea polyphenols before irradiation with UVB (4 kJ/m 2 ), lysates were prepared from the cells, and the immunoprecipitation was carried out using 4 g of Akt1/PKB␣ PH domain antibody or 3 g of anti-p70 S6-K antibody as described above. The enzyme-immune complex was washed three times with 0.5 ml of lysis buffer and once with 100 l of assay dilution buffer (20 mM MOPS, pH 7.2, 25 mM ␤-glycerophosphate, pH 7.0, 1 mM sodium orthovanadate, 1 mM DTT). For the Akt kinase assay, the enzyme immune complex was added to 10 l of assay dilution buffer, 10 M protein kinase A inhibitor peptide, 0.1 mM Akt substrate peptide, and 10 Ci of [␥-32 P]ATP, and for the p70 S6-K assay, it was added to 20 l of assay dilution buffer, 10 l of inhibitor mixture, 50 M S6 kinase substrate peptide, and 10 Ci of [␥-32 P]ATP. The reaction was incubated for 10 min at 30°C and centrifuged, and then 30 l of the supernatant fraction was transferred onto p81 phosphocellulose paper and allowed to bind for 30 s. The p81 papers were washed three times in 0.75% phosphoric acid and then washed once in acetone, and ␥-32 P incorporation was measured by scintillation counting.
Direct Inhibition of Akt and p70 S6-K Activities by Tea Polyphenols-The cells were treated with UVB (4 kJ/m 2 ) and lysed, and the immunoprecipitation was carried out using 4 g of anti-Akt1/PKB␣ PH domain or 3 g of anti-p70 S6-K antibody, and the enzyme-immune complex was washed with 0.5 ml of lysis buffer and with 100 l of assay dilution buffer as described above. For the Akt kinase assay, the enzyme immune complex was added to 10 l of different concentrations of tea polyphenols in assay dilution buffer, 10 M protein kinase A inhibitor peptide, 0.1 mM Akt substrate peptide, and 10 Ci of [␥-32 P]ATP, and for the p70 S6-K assay, it was added to 20 l of different concentration of tea polyphenols in assay dilution buffer, 10 l of inhibitor mixture, 50 M S6 kinase substrate peptide, and 10 Ci of [␥-32 P]ATP. The reaction and measurement were performed as described under "Akt and p70 S6-K Immunoprecipitation Kinase Assay." Statistical Analysis-Significant differences in the kinase activities were determined by using both Student's t test and Welch's t test.

Inhibition of UVB-induced PI3K Activity by EGCG or
Theaflavins-Exciting developments have recently implicated PI3K as an important factor in carcinogenesis (27)(28)(29)(30)(31)(32)(33)(34). Our previous studies have shown that PI3K is necessary for 12-Otetradecanoylphorbol-13-acetate-or EGF-induced cell transformation in mouse epidermal JB6 cells (35,36). Therefore, PI3K was also suggested to play a critical role in carcinogenesis induced by UVB. We first investigated the effect of the tea polyphenols, EGCG or theaflavins, on UVB-induced PI3K activity in JB6 Cl 41 cells. The results showed that pretreatment of cells with these polyphenols inhibits UVB-induced PI3K activity (Fig. 1). Although inhibition of UVB-induced PI3K activity by these polyphenols is suggested to be due to UVB absorption, our previous study showed that the attenuation of UVB-induced signaling by EGCG or theaflavins does not appear to result from UVB absorption (64).
Inhibition of UVB-induced Activation of Akt and p70 S6-K, Downstream Effectors of PI3K, by EGCG or Theaflavins-A number of studies have demonstrated important roles for Akt and p70 S6-K as downstream effectors of PI3K (38,40,47,51). Akt and p70 S6-K are suggested to be implicated in tumor promotion or carcinogenesis (27, 41-43, 51, 52). As shown in Fig. 2, pretreatment of cells with EGCG or theaflavins inhibited UVB-induced phosphorylation of Akt at both Thr 308 and Ser 473 , which are prerequisites for the catalytic activation of Akt (67)(68)(69). UVB-induced phosphorylation of p70 S6-K at Thr 389 and Thr 421 /Ser 424 was also blocked by pretreatment with these polyphenols (Fig. 2). In addition, pretreatment with these polyphenols attenuated UVB-induced activation of both Akt and p70 S6-K (Fig. 3). Phosphoinositide-dependent kinase (PDK-1) is known to phosphorylate Akt at Thr 308 , and it is also a downstream target of PI3K (24, 38, 40, 70 -72). PDK-1 is a constitutively active enzyme (73), but when the cells were treated with EGCG or theaflavins, its constitutive activation was not affected by these polyphenols (data not shown).
Signal Transduction of UVB-induced Activation of Akt and p70 S6-K-Our previous study demonstrated that EGCG and theaflavins inhibit UVB-induced phosphorylation of Erks and that theaflavins also attenuated UVB-induced phosphorylation of JNKs (64). Therefore, the inhibition of UVB-induced activation of Akt and p70 S6-K by these polyphenols may also be implicated in blocking Erks or JNKs. To determine the mechanism by which EGCG or theaflavins inhibit UVB-induced

FIG. 2. Inhibition of UVB-induced phosphorylation of Akt and p70 S6-K by pretreatment of cells with EGCG or theaflavins. JB6
Cl 41 cells (80% confluence) were starved by replacing the medium with 0.1% FBS MEM and culturing for 48 h. The cells were then pretreated with EGCG or theaflavins for 1 h at the indicated concentration. The cells were irradiated with UVB (4 kJ/m 2 ) and subsequently cultured for 30 min. The cells were lysed, and the phosphorylation levels were estimated by immunoblotting. activation of Akt and p70 S6-K, we investigated the signaling pathways involved in UVB-induced activation of kinases by using specific chemical inhibitors. Pretreatment of cells with a PI3K inhibitor, LY294002, blocked UVB-induced phosphorylation of Akt (at Thr 308 and Ser 473 ) and p70 S6-K (at Thr 389 and Thr 421 /Ser 424 ) (Fig. 4). LY294002 also inhibited UVB-induced activation of these kinases (Fig. 5), showing that in UVB signaling the activation of both Akt and p70 S6-K is mediated through PI3K. In addition, the phosphorylation and activation of Akt and p70 S6-K were attenuated by pretreatment of cells with a specific MEK inhibitor, U0126, or a specific p38 kinase inhibitor, SB202190 (Figs. 4 and 5), whereas the inhibition of phosphorylation and activation of these kinases by UVB was not observed in cells expressing a dominant negative mutant of JNK1 (Fig. 6) (74). Thus, Erks and p38 kinase, but not JNKs, mediate UVB-induced activation of Akt and p70 S6-K. Pretreatment with LY294002 restrained UVB-induced phosphorylation of Erks, suggesting that in UVB signaling the Erk pathway is mediated by PI3K. In agreement with this result, our previous study demonstrated that expression of a dominant negative mutant of the PI3K p85 subunit also inhibits UVBinduced phosphorylation of Erks, but not JNKs or p38 kinase (74). On the other hand, pretreatment with rapamycin, an inhibitor of mammalian target of rapamycin (mTOR) an upstream effector of p70 S6-K, inhibited UVB-induced phosphorylation and activation of P70 S6-K, whereas UVB-induced phosphorylation and activation of Akt was not affected by rapamycin (Figs. 4 and 5).
Direct Inhibition of UVB-induced p70 S6-K Activation by EGCG or Theaflavins-To test whether EGCG and theaflavins directly inhibit kinase activity of Akt and p70 S6-K induced by UVB, we performed kinase activity assays for Akt and p70 S6-K in the presence of these polyphenols. The enzymes were immunoprecipitated from cell lysate treated with UVB (4 kJ/ m 2 ) using an antibody against p70 S6-K or Akt. UVB-induced p70 S6-K activation was significantly blocked by the addition of EGCG or theaflavins, whereas these polyphenols only weakly inhibited UVB-induced Akt activation (Fig. 7). Thus, the inhibition of UVB-induced activation of p70 S6-K, but not Akt, by these polyphenols may be, in part, achieved by the direct blocking of kinase activity of p70 S6-K. DISCUSSION Green tea is widely used as a beverage in China, Japan, and other Asian countries, whereas black tea is more popular in Western countries (54,55,75). In recent years many animal studies and several epidemiological studies have reported the anti-carcinogenic effects of tea (54 -56, 59 -61, 75). The polyphenols from green tea and black tea, EGCG and theaflavins, respectively, are generally considered to be the most effective components for inhibition of carcinogenesis (56 -58, 75). Although several mechanisms explaining the anti-carcinogenic effects of tea have been presented (62,(75)(76)(77), we and others suggested that tea components target specific cell signaling pathways responsible for regulating cell growth, survival, and cell transformation (57,58,63,64). PI3K is an important regulatory protein that is involved in different signaling pathways and in the control of cell growth, survival, and malignant transformation (25)(26)(27)(28). PI3K forms complexes with some viral or cellular oncoproteins (29,30), which have transforming activities, and by itself also has oncogenic activity (31). The transforming effect of PI3K is suggested to be related to the activation of its downstream effector kinases, including Akt and p70 S6-K (27,37,(41)(42)(43)(51)(52)(53). We previously demonstrated that PI3K is necessary for 12-O-tetradecanoylphorbol-13-acetate-or EGF-induced cell transformation and AP-1 activation (35,36). We also recently found that rapamycin, which blocks p70 S6-K activation, inhibits EGF-induced cell transformation independently of AP-1 activation in mouse epidermal JB6 cells. 2 UVB is a crucial risk factor for skin cancer (3)(4)(5)(6) and also induces PI3K activation (20,21). Thus, PI3K and its downstream effectors are suggested to be important regulatory proteins in the tumor promotion effects of UVB.
In this study, we found that EGCG and theaflavins inhibit UVB-induced activation of PI3K and its downstream effectors, Akt and p70 S6-K. Furthermore, we demonstrated that the Erks and p38 kinase pathways are critical for UVB-induced activation of Akt and p70 S6-K. Fig. 8 shows a schematic of the proposed signaling model for Akt and p70 S6-K induced by UVB. Generally, Akt is believed to be regulated by a pathway mediated by PI3K and PDK-1 (24, 38, 40, 70 -72). In contrast, we recently demonstrated that UVB-induced Akt activation is mediated through Erks and p38 kinase-dependent mitogenand stress-activated protein kinase-1 rather than the PI3K/ PDK-1 pathway (74). That study (74) and the present study also suggested that PI3K is located upstream of Erks, because pretreatment of cells with a PI3K inhibitor, LY294002, or the expression of a dominant negative mutant of the PI3K p85 subunit inhibited UVB-induced phosphorylation of Erks, but not JNKs or p38 kinase.
In addition, we previously showed that overexpression of a dominant negative mutant of PKC/ (12) or pretreatment with antisense oligonucleotides of PKC specifically inhibited UVBinduced phosphorylation of Erks (78). PKC is regulated in vitro by PI-3,4,5-P 3 (79). PKC, as well as PKC, is activated in vivo through a pathway involving PI3K (80,81). Atypical PKCs (aPKCs), such as the PKC and PKC, have also been shown to be activated by PDK-1 (82,83), and the maximal phosphorylation and activation of aPKCs by PDK-1 require PI-3,4,5-P 3 . These findings suggested that in UVB signaling Erk activation is mediated by pathways involving PI3K/aPKCs or PI3K/PDK-1/aPKCs. On the other hand, PDK-1 is also suggested to serve as a Thr 229 kinase of p70 S6-K because the sequence surround- The cells were then pretreated with U0126, SB202190, LY294002, or rapamycin for 1 h at the indicated concentration. The cells were irradiated with UVB (4 kJ/m 2 ) and subsequently cultured for 30 min. A, the cells were lysed, and Akt or p70 S6-K was immunoprecipitated using a Akt1/PKB␣, PH domain antibody. The activity of Akt was assessed using Akt substrate peptide and [␥-32 P]ATP. Each bar indicates the mean Ϯ S.E. of at least two independent experiments. *, significant difference from the UVB control at p Ͻ 0.01. B, the cells were lysed, and p70 S6-K was immunoprecipitated using a p70 S6-K antibody. The activity of p70 S6-K was assessed using S6 substrate peptide and [␥-32 P]ATP. Each bar indicates the mean Ϯ S.E. of at least three independent experiments. *, significant difference from the UVB control at p Ͻ 0.01. The cells were then irradiated with UVB (4 kJ/m 2 ) and cultured for 30 min. The cells were lysed, and the activity of JNKs was determined as described under "Experimental Procedures." B, the transfectant cells were treated as described above and lysed. The levels of phosphorylation of p70 S6-K were estimated by immunoblotting. C, JB6 Cl 41 and dominant negative mutant of JNK1 cells were treated as described above and lysed. p70 S6-K was immunoprecipitated from the lysates using a p70 S6-K antibody. The activities of p70 S6-K were assessed using S6 substrate peptide and [␥-32 P]ATP. Each bar indicates the mean Ϯ S.E. of at least two independent experiments. ing the activation loop including Thr 229 of p70 S6-K is highly homologous to that of Akt, which is activated by PDK-1. Pullen et al. (73) reported that PDK-1 activates a p70 S6-K mutant with acidic residues at Thr 389 and four phosphorylation sites in the C terminus and that the catalytically inactive PDK-1 blocked insulin-induced p70 S6-K activation. However, the activation of p70 S6-K by PDK-1 was found to be independent of PI-3,4,5-P 3 or pretreatment with wortmannin or rapamycin (73). Romanelli et al. (84) showed that expression of a dominant negative mutant of PKC antagonized p70 S6-K activation by EGF and PDK-1. They also demonstrated that p70 S6-K forms complexes in vivo with PDK-1 and PKC. In addition, the interaction of the N-terminal region of p70 S6-K with the kinase domain of PKC and the requirement for interaction with the regulatory domain of PKC of the C-terminal region of p70 S6-K have been reported by Akimoto et al. (85).
We found that UVB irradiation causes translocation of PKC to cell membrane (78). Therefore, PDK-1 may be able to colocalize with aPKCs at the membrane. In this study, we showed that U0126, an inhibitor of MAP kinase/Erk kinase, repressed UVB-induced p70 S6-K phosphorylation at Thr 389 and Thr 421 / Ser 424 . Thus, the activation of p70 S6-K by PI3K is suggested to be mediated through interaction with aPKCs or phosphorylation at Thr 389 and Thr 421 /Ser 424 by PDK-1/aPKCs/Erk path-ways. Furthermore, although whether Akt phosphorylates mTOR is not clear (86,87), several studies have reported that phosphorylation and activation of mTOR is mediated by the PI-3/Akt pathway (88 -90). The Thr 389 residue of p70 S6-K is known to be rapamycin-sensitive, whereas the residues in the C terminus (like Thr 421 and Ser 424 ) are considered insensitive (91). However, more recent reports have questioned this difference (92,93). We also showed that rapamycin, an inhibitor of mTOR, not only represses phosphorylation of p70 S6-K at Thr 389 but blocks phosphorylation at Thr 421 /Ser 424 . In addition, SB202190, a specific p38 kinase inhibitor, inhibited UVB-induced p70 S6-K phosphorylation at Thr 389 and Thr 421 /Ser 424 and p70 S6-K has also shown to be phosphorylate by MAP kinases (94,95). These findings suggested that p38 kinase and Erks activate p70 S6-K directly by themselves or indirectly through other kinases including mTOR mediated by the mitogen-and stress-activated protein kinase 1/Akt pathway.
Our previous study showed that EGCG or theaflavins inhibit UVB-induced Erks phosphorylation (74), and the present study demonstrated that these polyphenols attenuated UVB-induced PI3K activation. Although the inhibition of Erks phosphorylation may be due to suppressed PI3K activation, the blocking of the Erks-mediated pathway was suggested to be involved in the inhibitory effects of Akt and p70 S6-K by tea polyphenols. On the other hand, EGCG and theaflavins also directly inhibited the activation of p70 S6-K but not Akt activation. Polyphenols are known to bind to proteins, as exemplified by their binding to the proline-rich salivary proteins (96 -98). These proteins have the general characteristics of a large size, a loose open structure, and a high proportion of hydrophobic amino acids and proline. Among the regulatory phosphorylation sites FIG. 7. Direct inhibition of UVB-induced p70 S6-K activation by EGCG or theaflavins. JB6 Cl 41 cells (80% confluence) were starved by replacing the medium with 0.1% FBS MEM and culturing for 48 h. The cells were irradiated with UVB (4 kJ/m 2 ) and subsequently cultured for 30 min. A, the cells were lysed, and Akt or p70 S6-K was immunoprecipitated using a Akt1/PKB␣, PH domain antibody. The activity of Akt was assessed using Akt substrate peptide and [␥-32 P]ATP with different concentrations of EGCG or theaflavins. Each bar indicates the mean Ϯ S.E. of at least three independent experiments. B, the cells were lysed, and p70 S6-K was immunoprecipitated using a p70 S6-K antibody. The activity of p70 S6-K was assessed using S6 substrate peptide and [␥-32 P]ATP with different concentrations of EGCG or theaflavins. Each bar indicates the mean Ϯ S.E. of at least three independent experiments. *, significant difference from the UVB control at p Ͻ 0.05.
FIG. 8. Effect of tea polyphenols on PI3K-dependent pathway induced by UVB irradiation. Exposure to UVB results in activation of the PI3K pathway. EGCG and theaflavins inhibit UVB-induced activation and phosphorylation of PI3K and its downstream effectors, Akt (Thr 308 /Ser 473 ) and p70 S6-K (Thr 389 and Thr 421 /Ser 424 ). The tea polyphenols did not affect phosphorylation of Akt (Thr 308 ) by PDK-1. UVBinduced p70 S6-K activation is also blocked by EGCG or theaflavins. The Erk kinase pathway, critical for UVB-induced activities of Akt and p70 S6-K, is also inhibited by tea polyphenols. 1, activation; Ќ, inhibition; tea bags, tea polyphenols. that are implicated in the activation of p70 S6-K, Ser 404 is surrounded by large bulky hydrophobic amino acids in the Ϫ1 and ϩ1 positions, similar to Thr 229 and Thr 389 , and the remaining phosphorylation sites in C-terminal are followed by proline at the ϩ1 position, similar to Ser 371 (99). Therefore, EGCG and theaflavins may modulate the conformation of p70 S6-K by binding to the hydrophobic amino acids described above (Fig.  7B). In contrast, the binding of these polyphenols to Akt was suggested to be unnecessary or insufficient for inhibition of its activation (Fig. 7B).
In summary, the present results show that EGCG and theaflavins inhibited the activation of PI3K and also attenuated the activation of Akt and p70 S6-K, downstream effectors of PI3K, by inhibiting PI3K and Erk activation in UVB signaling. Furthermore, these polyphenols directly blocked UVBinduced p70 S6-K activation. Because PI3K and its downstream effectors are considered to play a critical role in carcinogenesis, these results provide insight into the biological actions of tea polyphenols on UV-induced carcinogenesis and on the molecular basis for the development of new chemopreventive agents.